物理 · Physics

The Standard Model

Day 25 · 2026 · Phase F Particles & the Standard Model
A table that fits on a T-shirt lists every ingredient in your body. It is absurdly precise and visibly incomplete — both at once.
Take matter apart layer by layer: molecules, atoms, nuclei, protons and neutrons, quarks. There the trail ends — look closer and all you find is a point with no structure. What's left is a dozen particles and four forces, three of which fit into one framework while the fourth (gravity) still waits outside. This table predicts the electron's magnetism to twelve decimal places, yet cannot say why there are exactly three generations, or why the masses span twelve orders of magnitude. The most successful theory in history is also the one that makes physicists most uncomfortable.

The Inventory of Matter

Fermions · 1897–2000
Intuition The periodic table has 118 elements, which sounds like a lot of variety. Building every one of them takes just three particles: the up quark, the down quark, and the electron. A proton is two ups and a down, a neutron is two downs and an up, electrons queue up outside, and that's all of chemistry. Add the electron neutrino, which barely interacts with anything, and your world is complete — four particles.
Mechanism Matter particles are called fermions, in two families: quarks feel the strong force and are locked inside protons and neutrons; leptons (the electron and neutrino family) don't feel it and travel alone. Each "generation" is two quarks plus two leptons — three generations, twelve particles, twenty-four counting antiparticles. Every fermion obeys the Pauli exclusion principle: two identical fermions cannot occupy the same state. That single rule is why you don't sink into your chair and why white dwarfs don't collapse.
Gen 1 Gen 2 Gen 3 Force carriers Gives mass Missing Quarks Leptons uct dsb eμτ νe νμ ντ gγ WZ H upcharmtop downstrangebottom electronmuontau e neutrinoμ neutrinoτ neutrino gluon · strongphoton · EM W · weakZ · weak Higgs 2.2 MeV1.27 GeV173 GeV 4.7 MeV93 MeV4.18 GeV 0.511 MeV105.7 MeV1.777 GeV under 1 eVunder 1 eVunder 1 eV mass 0mass 0 80.4 GeV91.2 GeV 125 GeV Gravity this table can't hold it Quarks · feel the strong force Leptons · don't Bosons · carry forces Everyday matter uses only the left column: up, down, electron, electron neutrino Masses run from under 1 eV to 173 GeV — twelve orders of magnitude, unexplained
The three columns are copies of each other: identical properties, just heavier and shorter-lived as you move right.
The counterintuitive part "Only three generations" was counted, not derived. The more things a Z boson can decay into, the faster it dies and the wider its resonance peak. Measure that peak shape precisely and you can read off how many invisible exits it has — including neutrinos nobody can see. LEP counted 2.984 ± 0.008, pinning it cleanly at 3. Theory offers no account of that 3 whatsoever: it reads like a note copied down from the experiment.
Cross-disciplinary reading · Chemistry / Archaeology / Medicine
  • Chemistry: the entire periodic table is one generation of particles rearranged. Mendeleev's periodicity was once the deepest empirical law in nature; today it is a corollary of "three particles plus Pauli exclusion" — a top-level law demoted to a mid-level consequence, which in science is usually what progress looks like.
  • Archaeology: second-generation muons rain down on your head from cosmic rays every minute. In 2017 an international team put muon detectors inside the Great Pyramid and, by comparing how many muons arrived from each direction, imaged a thirty-metre void through a hundred metres of stone.
  • Medicine: a PET tracer emits positrons, which annihilate with electrons in your body into two photons flying back to back; the scanner locates the source along that line. Antimatter is hospital equipment booked by the hour, not a science-fiction prop.
One line: all the matter in your body uses only the leftmost column, and nobody knows why the other two exist.
Think: Second- and third-generation particles live less than a microsecond. On what grounds do we say they "exist"?
They are made over and over in colliders, with masses, lifetimes and decay channels measured to high precision; and leave them out and many precision low-energy calculations stop matching. In physics the criterion for existence is a reproducible causal effect, not whether you can hold it.

Forces Are Exchanges

Gauge bosons · Yukawa 1935
Intuition Two people on ice throwing a ball back and forth drift apart — a force isn't an invisible hand reaching across a gap, it's something being swapped. The particles that do the swapping are bosons, and their temperament is the opposite of a fermion's: they are happy to pile into the same state (a laser is a mob of photons marching in step). Photons carry electromagnetism, gluons the strong force, W and Z the weak force.
Mechanism How far a force reaches is written in its carrier's mass. Conjuring a carrier of mass m costs mc2 of energy, the uncertainty relation only lets it live for a moment, and the farthest it gets is
R ≈ ħmc
R is the range, "how far this force reaches"; ħ is read h-bar, Planck's constant divided by 2π, the conversion unit of the quantum world; m (letter m) is the carrier's mass; c is the speed of light. m sits in the denominator: heavier carrier, shorter reach. If the carrier's mass is 0 (the digit zero), the denominator vanishes and the range is infinite.
The photon is massless, so electromagnetism falls off as 1/r2 all the way to the horizon — which is why starlight can travel for billions of years into your eye. The W boson weighs 80.4 GeV, giving about 2.5×10−18 metres, a thousand times smaller than a proton. Much of what we call the "weakness" of the weak force is simply that it can't reach you. The gluon is massless too and ought to reach forever, but gets cut off at nuclear scale for a different reason.
Swap a boson = feel a force boson fermion fermion time runs upward · both ends recoil Range is written in the carrier's mass Electromagnetic · photon, mass 0 Infinite — starlight crosses galaxies to your eye Strong (residual, in nuclei) · gluons About 10⁻¹⁵ m — precisely one nucleus Weak · W and Z, a hefty 80–91 GeV About 10⁻¹⁸ m — a thousandth of a proton (bar lengths are schematic and logarithmic)
Left: two fermions swap a boson and thereby feel a force. Right: the heavier the carrier, the shorter the reach.
The counterintuitive part The "weakness" of the weak force is an artefact of scale. Push above 100 GeV and the weak and electromagnetic forces become comparable in strength — they were one force to begin with (the electroweak interaction), split into two faces by the Higgs field in today's cold, low-energy universe: one face with a massless photon reaching to infinity, the other dragging heavy W and Z bosons and stuck inside nucleons. We call it weak only because we live in the cold.
Cross-disciplinary reading · Chemistry / Computer science
  • Chemistry: neutral molecules still attract one another (van der Waals forces) because fluctuations in their charge distributions leak a little residual electromagnetism. The nuclear force gluing protons and neutrons together works the same way — a leftover of the colour force between quarks. A gecko on a ceiling and a nucleus that doesn't fly apart are the same "residual force" trick played at two scales.
  • Computer science: in a graph neural network no node acts at a distance; influence travels along edges as messages, and the number of rounds sets the "range" (the receptive field). Gauge theory is the physical version of the same statement: an interaction is not a formula written between two particles, it is a field that must be handed along point by point.
One line: how strong a force is and how far it reaches are both encoded in the mass of its messenger.
Think: If virtual particles "borrow" energy, is conservation of energy violated?
No. A virtual particle isn't a real particle that briefly exists; it's an internal line in a perturbative calculation, a bookkeeping term. Energy is exactly conserved from initial to final state — only that internal line is exempt from the free-particle relation between energy and momentum. "Borrowing energy" is a figure of speech, not live commentary.

Why No One Ever Holds a Quark

Quantum chromodynamics · asymptotic freedom 1973
Intuition Familiar forces weaken with distance: gravity and electromagnetism both fall off as 1/r2. The colour force does the opposite — the further you pull, the harder it gets. Separating a pair of quarks is like stretching a rubber band; at some length the energy you have poured in is enough to create a fresh pair of quarks, and the band snaps. A new quark appears at each broken end, you are left holding two particles, and still not one free quark.
Mechanism Quarks carry something called colour charge, in three varieties conventionally labelled red, green and blue — pure labels, nothing to do with visible colour. Here is the decisive difference: photons carry no electric charge, but gluons do carry colour charge. Charged messengers tangle with each other, so the field lines stop spreading out and get squeezed into a narrow tube. The tube's energy grows linearly with length, at a tension of roughly 1 GeV per femtometre. Run it the other way and the coupling weakens at very short distances, leaving quarks nearly free — asymptotic freedom (Nobel Prize 2004).
Pull harder and you only make more particles quark antiquark ① Bound together a colour flux tube between them ② Pull: the energy climbs linearly about 1 GeV per femtometre stretched a new pair pops out at the break ③ The tube snaps two bound states, still no free quark The energy you invest turns into new particles rather than releasing the quark — that is confinement
The tube's tension doesn't fade with distance, so there is always enough energy to make particles before there is enough to break free.
The counterintuitive part The Higgs accounts for under 1% of your body weight. A proton weighs 938 MeV, while the three quarks inside it add up to about 9 MeV. The other 99% is the energy of the gluon field plus the kinetic energy of quarks rattling around — confined to a box a femtometre across, they move ferociously fast. E = mc2 here is not a metaphor for bombs, it is a literal translation: most of what the bathroom scale reads is not "amount of stuff" but imprisoned energy.
Cross-disciplinary reading · Complex systems / Computational science
  • Complex systems: "the whole is more than the sum of its parts" becomes a weighable number here — proton 938, parts 9. The excess is the interaction itself. The thread running through "More Is Different" is usually only discussable in qualitative terms; here it has a precise instance.
  • Computational science: the colour force is so strong that perturbative expansion fails outright, leaving no option but to chop spacetime into a four-dimensional lattice and integrate by Monte Carlo on a supercomputer. In 2008 one collaboration computed the masses of the proton, the neutron and their relatives from first principles this way, matching experiment to a few percent — a literal calculation of your body weight from the ground up.
One line: most of your weight isn't matter, it's imprisoned energy.
Think: If quarks can never get out, how do we know they're in there?
Fire high-energy electrons at a proton and the angular distribution of the scattering shows they are hitting three point-like hard cores, not a uniform smear. Colliders also show "jets" — a struck quark shatters into a tight cone of hadrons whose direction and energy faithfully record the momentum of the original quark.

Quirks of the Weak Force, and the Edges of the Table

Parity violation 1957 · 19 free parameters
Intuition Gravity, electromagnetism and the strong force all play fair with mirrors: the mirror-image process is equally legal in reality. The weak force is not. In 1956 Chien-Shiung Wu cooled cobalt-60 to near absolute zero, lined up the nuclear spins with a magnetic field, and counted which way the emitted electrons flew — they clearly favoured one side. The mirrored version is something nature never performs. At the most basic level, the universe can tell left from right.
Real experiment Mirror world mirror nuclear spin Co-60 Electrons mostly fly against the spin spin unchanged Co-60 Mirror prediction: mostly along the spin Nature has never done this
A mirror flips momentum but not spin, so the two pictures make distinguishable predictions — and the experiment sides with the left one.
Mechanism The weak force is also the only one that can change "flavour": only it can turn an up quark into a down quark, or a muon into an electron. The first step of sunlight is exactly this — two protons collide and one becomes a neutron. Nothing but the weak force can do it, and it is absurdly slow, which is why the sun smoulders for ten billion years instead of detonating. The price is that the table is not self-contained: the Standard Model has 19 numbers that can only be filled in by measurement (nine fermion masses, three coupling constants, four quark-mixing parameters, two Higgs parameters and one strong-CP angle). Add the neutrino masses and mixings discovered later and the count rises to around twenty-six.
The counterintuitive part Neutrino mass has, strictly speaking, already falsified a small piece of the original Standard Model, which explicitly assumed neutrinos were massless. In 1998 Super-Kamiokande saw atmospheric neutrinos change flavour in flight, forcing masses to be non-zero (Nobel Prize 2015). Patching mass in is easy; the trouble is that the patch isn't unique — add right-handed neutrinos, or let the neutrino be its own antiparticle? The choice bears directly on why the universe has more matter than antimatter. This crack opened from the experimental side.
Cross-disciplinary reading · Biology / AI
  • Biology: life on Earth uses almost exclusively left-handed amino acids and right-handed sugars. Since the weak force distinguishes left from right by nature, many have guessed that's where handedness came from. The honest state of play: the energy difference the weak force imposes on a pair of mirror molecules is around 10−17 in relative terms, has never been measured directly, and mainstream accounts still credit random symmetry breaking plus autocatalytic amplification. An asymmetry at the bottom does not make it the cause of yours.
  • AI: a rough gauge of a theory's "unexplained fraction" is how many of its parameters can only be fitted. All 19 Standard Model numbers are read in from experiment and none can be derived — which makes it look more like an extremely well-fitted model than a principle anyone has understood.
One line: the Standard Model isn't a theory of everything, it's an outrageously precise inventory with its own boundary written on the label.
Think: Missing gravity, stuffed with free parameters — on what grounds is it called "successful"?
Because inside its jurisdiction it has never been wrong. Theory and experiment for the electron's magnetic moment agree to one part in 10¹², the most precise comparison in all of science; and the particles it predicted before they were found line up in a row: W, Z, top quark, Higgs. Success here means "never lost where it had authority", not "explained everything".

Going deeper

Why exactly three generations?
There is no answer, only constraints. The Z width at LEP rules out a fourth light neutrino, and Higgs production rates make a fourth generation very unlikely. One tantalising clue: for the quark mixing matrix to accommodate a complex phase — and thus for matter and antimatter to behave differently — at least three generations are required. That is how Kobayashi and Maskawa inferred a third generation back in 1973, when the second wasn't even complete. But it only shows three are necessary, not why there aren't more.
What actually is a "particle" in this table?
Not a tiny ball — an excitation of a field, which is the whole point of quantum field theory. Each row corresponds to a field filling the entire universe, and a particle is one indivisible quantum of vibration in it. Identity falls out for free: all electrons are exactly alike because they are the same excitation of the same field. Read each cell as "the universe contains this field".
Is the muon g−2 "anomaly" still a hint of new physics?
It was once the hottest candidate; today it needs a heavy discount. Fermilab pushed the measurement to the parts-per-billion level, but the theory side moved further still: the hadronic contribution from lattice QCD disagreed with the older data-driven method, and adopting the former brings prediction and measurement into broad agreement. The concrete lesson: in a claimed gap between experiment and theory, sometimes the theoretical uncertainty is the main character.

Further reading